be0b90eed394fb4cbad3233ab06d3c66

November 2006
A Dynamic Perspective on Government

Broadband Initiatives
By Jerry Ellig
Project Director: Adrian T. Moore
POLICY
STUDY
349
Reason Foundation
Reason Foundation’s mission is to advance a free society by develop-
ing, applying, and promoting libertarian principles, including indi-
vidual liberty, free markets, and the rule of law. We use journalism and
public policy research to influence the frameworks and actions of poli-
cymakers, journalists, and opinion leaders.
Reason Foundation’s nonpartisan public policy research promotes

choice, competition, and a dynamic market economy as the founda-
tion for human dignity and progress. Reason produces rigorous, peer-
reviewed research and directly engages the policy process, seeking
strategies that emphasize cooperation, flexibility, local knowledge, and
results. Through practical and innovative approaches to complex prob-
lems, Reason seeks to change the way people think about issues, and
promote policies that allow and encourage individuals and voluntary
institutions to flourish.
Reason Foundation is a tax-exempt research and education organiza-

tion as defined under IRS code 501(c)(3). Reason Foundation is sup-
ported by voluntary contributions from individuals, foundations, and
corporations. The views are those of the author, not necessarily those
of Reason Foundation or its trustees.
Copyright © 2006 Reason Foundation. Photos used in this publication

are copyright © 1996 Photodisc, Inc. All rights reserved.
Reason Foundation
A Dynamic Perspective on Government

Broadband Initiatives
By Jerry Ellig
Project Director: Adrian T. Moore
Executive Summary
D ebate over government provision of broadband has generated many of the usual arguments
over the pros and cons of government service provision. On the one hand, such initiatives
might make broadband more affordable and hasten its adoption. On the other hand, they could also
generate significant costs for taxpayers and stunt incentives for cost containment. Such arguments
commonly occur when governments consider direct provision of electricity, gas, water, roads, and
many other services that tend to be provided by monopolies that invest in long-lived assets. Less
extensively discussed, however, are some unique challenges that arise because broadband is anew,
fast-changing technology available from competing suppliers. Policymakers need to consider some
unique problems when a government enterprise enters a dynamic market such as the provision of
Internet services.
Traditionally, infrastructure like water systems, gas distribution and electric distribution has
involved a fixed investment that was very large compared to the ongoing operating cost. The
technology of the infrastructure itself changed relatively slowly. As a result, local governments
could usually invest in what appeared to be the best technology at the time without having to worry
much about whether they chose the right technology. Pricing and selling the service was relatively
easy. Most people need water, heat and electricity, and the service providers usually had a
monopoly. Long asset lives and slow technological change made long depreciation schedules
possible. Service providers could be reasonably confident of recovering their capital costs over a
long time period.
However, these static monopolies differ significantly from services like Internet provision, which
are dynamic due to their fast-changing technology and variety of services. Unlike heat, water and
electricity, high-speed Internet service is not viewed as a need by many people. Companies
continually change their offerings and prices to appeal to a variety of consumer desires. This
dynamic competition upsets the tranquil conditions that make government provision of a service
easier.
Scholarship on dynamic competition suggests seven new issues that are likely to be significant in
municipal provision of Internet service:
Competition: Unlike a monopolist, an enterprise that faces competition cannot count on a

captive market. The only exceptions might be small communities serviced only by expensive
alternatives, or municipalities willing to commit to very large subsidies for their broadband
systems.
Performance Competition: Competitive businesses seek to continually improve performance—

or even develop new aspects of performance that were not previously thought capable of
improvement. Comparing prices and services offered by government-sponsored Internet
provision to those in the private sector, the prices and performance of existing government
systems are often inferior to those of existing private systems.
Continuous Improvement: One indicator of the extent of change is the pace at which prices of
goods and services fall as technology improves, costs fall, or competition intensifies. Unlike in
state-owned utilities, this has occurred frequently in the market for Internet service, as well as
in related or analogous markets such as wireless communications, telephone equipment, and
telecommunications services.
Technological Change and Lock-In: “Lock-in” occurs when an initial decision gives one
technology a slight edge, then sets in motion a process which leads that technology to dominate
the market. The market can get locked in to an inferior technology due to the decisions of the
early adopters, and subsidies may prompt early adopters to choose an inferior technology.
Government broadband plans should squarely address the potential for lock-in and explicitly
evaluate whether subsidies would give an inferior technology an artificial boost.
Obsolescence: Wireless technology improves rapidly, and as a result capital investment

becomes obsolete more quickly. Business plans for government broadband enterprises need to
assume faster depreciation rates, and concomitantly higher prices, than have traditionally been
used for government utilities. For example, a workable plan for municipal Wi-Fi needs to
assume that revenues will not just cover operating costs plus interest, but also recover the
initial capital outlay in three to five years.
Risk: Financially, investment in a dynamic field such as Internet provision is less of a “sure
thing” than a conventional government monopoly such as electricity, gas or water provision.
Decision-makers must ensure that governments do not finance broadband as if it were a
traditional low-risk investment in infrastructure, as some have done, so that spending decisions
are weighed appropriately.
Uncertainty: The fact that uncertainty affects private business shareholders’ financial fortunes
gives them strong incentives to seek out management that will exercise sound judgment. For
government broadband enterprises, taxpayers bear the uncertainty in their role as the ultimate
owners. At a minimum, therefore, effective accountability requires that government broadband
initiatives should have accountability and transparency for taxpayers at least as good as that
which publicly held companies must have for their shareholders.
While many broadband initiatives require some type of public sponsorship or investment—either
by government or by government-owned entities—a recent twist appears to offer the public a much
better deal. In some cities, such as Philadelphia and San Francisco, private firms have proposed to
build Wi-Fi networks at no cost to taxpayers. There is nothing inherently wrong with proposals for
free or privately subsidized Wi-Fi. However, governments need to realize that rights-of-way and
light poles are valuable assets, and access to these assets would bestow a significant competitive
advantage on any firm selected to use them.
Any local government that grants one Wi-Fi provider an exclusive right to use right-of-way and
poles risks distorting competition in whatever markets are generating the revenue stream that will
subsidize the Wi-Fi service. A monopoly that gives away Wi-Fi to build demand for other services
it might sell to Wi-Fi users might be able to charge a higher price for these other services than it
would in the presence of other Wi-Fi competitors. For this reason, local governments should
beware of granting one Wi-Fi provider exclusive access to public assets, even if the Wi-Fi service
itself is free of charge to users. At a minimum, decision-makers should assess whether exclusive
access would distort competition in the markets for other goods and services sold by the Wi-Fi
company.
The factors outlined above need not imply that government-provided broadband is a bad idea.
However, no plan for government-sponsored broadband should be considered complete or
responsible unless it addresses many factors. Government faces the daunting challenge of entering
a market where technological change is swift, the future is uncertain, and competitors’ actions are
unpredictable—a playing field fundamentally different from the stable, predictable utility markets
that have traditionally attracted public investment.
R e a s o n F o u n d a t i o n
Table of Contents
Introduction...............................................................................................................1
What Is Dynamic Competition and Why Does It Matter?............................................3
New Issues for Decisionmakers ..................................................................................6
A. Competition....................................................................................................................... 6
B. Performance Competition ................................................................................................ 11
C. Continuous Improvement ............................................................................................... 14
D. Technological Change and Lock-In ................................................................................. 16
E. Obsolescence.................................................................................................................. 18
F. Risk ................................................................................................................................. 19
G. Uncertainty .................................................................................................................... 22
“Free” Networks: Beware of Geeks Bearing Gifts......................................................24
Conclusion...............................................................................................................27
Private and Public Broadband Providers ...................................................................29
Notes on Sources and Calculations...........................................................................33
About the Author .....................................................................................................34
Related Reason Studies ............................................................................................35
Endnotes..................................................................................................................36
DYNAMIC PERSPECTIVES ON GOVERNMENT BROADBAND 1
Part 1
Introduction
P roposals for government provision of broadband Internet service have generated significant
debate in many localities. Such proposals come in many forms. Philadelphia plans to have
Earthlink build a wireless (Wi-Fi) network that would offer service for about $20 per month, or
$10 for low-income users.1 In June 2005, voters in Lafayette, Louisiana approved a bond issue to
build a fiber network.2 In November 2005, voters in 17 Iowa cities voted in favor of creating
telecommunications utilities that could offer phone and Internet service.3 A 2004 survey found that
621 public power systems supply broadband services. About 23 municipalities offer fiber optic
service to the public.4 News reports claim that “hundreds of cities” are considering some type of
municipal Wi-Fi.5 San Francisco Mayor Gavin Newsome even declared free or low-cost wireless
Internet access a “fundamental right.”6 Google then grabbed headlines by proposing to turn San
Francisco into a Wi-Fi hotspot at no cost to the city. Media reports state that “many cities are
approaching Wi-Fi as fundamental infrastructure akin to electricity and water.”7
Debate over government provision of broadband has generated many of the usual arguments over
the pros and cons of government service provision. On the one hand, such initiatives might make
broadband more affordable and hasten its adoption. On the other hand, they could also generate
significant costs for taxpayers and stunt incentives for cost containment. Such arguments
commonly occur when governments consider direct provision of electricity, gas, water, roads, and
many other services that tend to be provided by monopolies that invest in long-lived assets.
Less extensively discussed, however, are some unique challenges that arise because broadband is a
new, fast-changing technology available from competing suppliers. Dynamic competition is
different from static monopoly. Policymakers need to consider some unique and different
problems when a government enterprise enters a dynamic market. Dynamic competition could
have a significant effect on the success or failure of a government broadband enterprise. The
presence of dynamic competition raises a rather different series of questions that thoughtful
decisionmakers should ask before launching government-sponsored broadband initiatives.
Several types of government-sponsored broadband initiatives have evoked discussion in recent

years. Some municipal utilities offering cable service offer cable modems, just like private cable
companies. The physical plant consists of coaxial cable or a combination of fiber optic and coaxial
cable. Others offer fiber optic cable all the way to the home, much like some of the phone
companies are starting to build. Some municipal electric utilities will no doubt seek to follow the
2 Reason Foundation
example of Manassas, Virginia, in providing broadband over power lines.8 Most recent
controversy has swirled around municipal proposals for “Wi-Fi.” Wi-Fi operates much like a
cordless telephone, providing computers with cordless Internet access at distances up to 300 feet
from Wi-Fi transmitters. Like cordless telephones, Wi-Fi uses unlicensed radio spectrum. Some
municipalities, such as Philadelphia and San Francisco, seek to blanket a large area with Wi-Fi
transmitters to create a citywide Wi-Fi network. Finally, a longer-range wireless technology called
“WiMAX” is frequently discussed but still under development. WiMAX might provide cordless
Internet access up to 25 miles from transmitters.9
Part 2
What Is Dynamic Competition and

Why Does It Matter?
D ynamic competition involves two factors: dynamism and competition. This may seem
obvious, but it means that dynamically competitive markets are substantially different from
the relatively static monopolies that characterize many government infrastructure investments.
Traditionally, infrastructure like water systems, gas distribution, and electric distribution has
involved a fixed investment that was very large compared to the ongoing operating cost. The
technology of the infrastructure itself changed relatively slowly. As a result, local governments
could usually invest in what appeared to be the best technology at the time without having to worry
much about whether they chose the right technology. Pricing and selling the service was relatively
easy. Most people need water, heat, and electricity, and the service providers usually had a
monopoly. Long asset lives and slow technological change made long depreciation schedules
possible. Service providers could be reasonably confident of recovering their capital costs over a
long time period. Pressures to replace the monopoly or allow competition were often minimal,
because there were no radically new or different technologies out there that offered additional
obvious and significant benefits. Government-owned utilities had a good shot at breaking even
and providing at least an acceptable level of service, if that’s what local officials wanted.10
Dynamic competition upsets this tranquil situation. By far the most prominent dynamic concept of
competition is associated with economist Joseph Schumpeter. Schumpeter argued that the most
significant advances in human wellbeing are triggered by “competition from the new commodity,
the new technology, the new source of supply, the new type of organization—competition which
commands a decisive cost or quality advantage and which strikes not at the margins of the profits
and the output of existing firms, but at their foundations and their very lives.”11 Journalists
observing the explosion of new information and communications technologies have explicitly
noted Schumpeter, with headlines declaring that “Today’s Hottest Economist Died 50 Years
Ago.”12
In addition to Schumpeter, a variety of other scholars have also developed dynamic theories of
competition.13 In “evolutionary” competition theories, a lot of things happen that most people
would readily identify as realistic features of real-world competition. Different firms have
4 Reason Foundation
different abilities, novelty constantly arises, innovation occurs as firms learn, and there are limits to
the amount of information decisionmakers can acquire and process. Competition is an open-ended
process of innovation, experimentation, and feedback.14 Some of Schumpeter’s fellow “Austrian”
economists developed complementary theories that emphasize competition as a process for the
discovery of new knowledge. Real-world competition is a dynamic process of trial and error. The
purpose of competition is to reveal what services, costs, and prices are possible. The firms that
survive and grow are those that do a better job than others of anticipating what consumers want
and finding the best way to produce it. 15 More recently, the interaction of competition and
technological change has led scholars to examine whether small variations in initial conditions can
lead to large and unpredictable changes in the market’s evolution.16
Finally, strategic management scholars explicitly view competition as a continual striving to cost-
effectively develop superior capabilities to serve consumers.17 Strategic management emphasizes
the importance of ensuring that a firm’s capabilities are a good match to its environment. In a
dynamically competitive market, some of the most important capabilities are the ability to
innovate, to change business strategy rapidly, to drop and add services in response to customer
needs, to upgrade products with new technology and features, and to change prices as market
conditions change.
Dynamic competition is a powerful force in the marketplace that makes a substantial

contribution to improvements in human wellbeing.
The U.S. experience with regulation and deregulation of infrastructure industries reveals
substantial evidence that dynamic competition is a powerful force in the marketplace that makes a
substantial contribution to improvements in human wellbeing.
Empirical studies of deregulated industries consistently find that competition generates more
consumer benefits than economists predicted based on pre-deregulation costs and market
conditions:18
Airline deregulation led to more frequent flights, more nonstop flights, and more possible
connections on heavily traveled routes. One of the most authoritative studies pegged the value
of increased flight frequency at $10.3 billion annually in 1993 dollars.19 Spurred largely by the
hub-and-spoke route system, the increase in flight frequency was a surprise; prior to
deregulation, researchers predicted that airlines would cut costs by reducing flight frequency.20
Another substantial innovation in the airline industry came from the low-cost carriers, such as
Southwest. Low-cost airlines cut back on frills, make more efficient use of labor, and avoid
unnecessary expenses. Southwest, originally an intrastate carrier in Texas, could not compete
with the established national airlines prior to deregulation, because the federal government
refused to approve any new airlines during the 40 years prior to 1978.21
AT&T lagged in introducing fiber optic cable and digital switching until it faced substantial
competition from MCI and Sprint. In 1985, for example, MCI and Sprint combined had 1.5
times the fiber mileage of AT&T; by 1994, AT&T had 1.3 times more fiber than MCI and
Sprint combined.22
Under traditional monopoly regulation, residential gas customers bear the risk of fluctuations
in natural gas prices. In states that permit consumers to choose their gas supplier, consumers
can reduce this risk if they choose a supplier who offers a fixed-price contract.23 Suppliers
usually hedge this risk in gas futures markets, thus passing it on to speculators.
Increased competition resulting from trucking deregulation spurred innovation by creating
pressure to reduce costs and increase the reliability of service. Carriers can now monitor
vehicle temperatures, security, and weight through remote communications technologies. For
less-than-truckload shipments, freight forwarders provide customers with a more precise
matching of transportation services with their needs. Within five years of interstate trucking
deregulation in 1980, shippers saved nearly $6-10 billion annually in lower rates and $1 billion
annually due to more rapid and reliable service.24
Actual rate and cost reductions resulting from railroad deregulation were two or three times the
size that economists predicted prior to deregulation. In addition, deregulation led to significant
improvements in service quality that no one predicted beforehand. Researchers at the
Interstate Commerce Commission and Federal Trade Commission estimated that more reliable
rail service reduced manufacturers’ inventory costs by $5-10 billion annually in 1987.25 A
Brookings Institution study estimated that speedier rail service made shippers better off by $2
to $6 billion in 1977 dollars.26
If even “low-tech” industries like railroads experience significant effects from dynamic
competition, it surely deserves attention when policymakers consider investing tax dollars in fast-
changing communications technology.
6 Reason Foundation
Part 3
New Issues for Decisionmakers
D ynamic competition complicates the assessment of government-sponsored broadband

enterprises. Scholarship on dynamic competition suggests seven new issues that are unlikely
to be significant under static monopoly: Competition, Performance Competition, Continuous
Improvement, Technological Change and Lock-in, Obsolescence, Risk, and Uncertainty. There is
substantial evidence from broadband, telecommunications, and other network industries that these
dynamic issues are not just theoretical possibilities, but real concerns that could have a substantial
effect on the success of a government broadband provider.
A. Competition
Unlike a monopolist, an enterprise that faces competition cannot count on a captive market. In
many cases, government-sponsored broadband will have to compete with incumbent firms, such as
cable, telephone, and wireless companies that already have a substantial head start. Judging from
the money, time, and effort these firms have spent fighting government-sponsored broadband, they
clearly view it as a competitor. Because broadband providers rarely have a captive market, the
“penetration rates” and profit margins for a government-sponsored broadband service will likely be
lower than for a municipal electric or water company.
The penetration rate is the percentage of eligible or targeted customers who actually choose to
subscribe to the service. It is different from market share, which is the percentage of customers
using the service who get it from a particular provider. In 2005, broadband was available to more
than 90 percent of U.S. households, yet only about 32 percent subscribed.27
Nevertheless, government broadband initiatives often assume relatively high penetration rates and
market shares. A 14-city fiber network planned in Utah, for example, assumes a 50 percent
penetration rate after five years and a 55 percent penetration rate after 10 years, despite the
presence of competitors.28 The plan for Lafayette, Louisiana’s fiber network assumes residential
cable TV penetration of 50 percent after 2008 and residential Internet penetration of 27 percent by
2009, rising to 37 percent in 2027. It also assumes that margins (earnings before interest, taxes,
depreciation and amortization) will exceed 50 percent over the long term—a highly unlikely
assumption for a capital-intensive system that faces competition from similar competitors.29 The
financial assumptions underlying Philadelphia’s proposed wireless network imply a residential

penetration rate of 13 percent in the first year, rising to 22 percent in year five. Though apparently
more modest than the assumptions in other cities’ fiber optic plans, this figure still implies that the
city’s wireless network would acquire all new broadband customers in the city during its first year,
plus take 5.5 percent of the market from Verizon and Comcast, the incumbents already serving
Philadelphia. Despite the presence of well-funded competitors, the Philadelphia wireless plan
assumes margins of 30-42 percent.30
Despite the presence of well-funded competitors, the Philadelphia wireless plan assumes
margins of 30-42 percent.
Are these kinds of penetration rates realistic? The evolution of broadband market shares over the
past six years might shed some light on that question. The FCC collects data on residential and
small business use of “high-speed” lines (capable of more than 200 kbps at least one-way). The
accompanying figures and tables hold several lessons for any potential competitor.
Figure 1: Residential and Small Business High-Speed Lines
Source: FCC High-Speed Services Report (July 2006), Table 1.

8 Reason Foundation
Table 1: Subscriptions to High-Speed Services (Year-End)

1999 2000 2001 2002 2003 2004 2005
Fiber/powerline 312,204 376,203 494,199 548,471 602,197 697,779 454,055
Other wireline 609,909 1,021,291 1,078,597 1,216,208 1,305,070 1,468,566 876,286
Satellite/wireless 50,404 112,405 212,610 276,067 367,118 549,621 3,809,247
DSL 369,792 1,977,101 3,947,808 6,471,716 9,509,442 13,817,280 19,514,318
Cable 1,411,977 3,582,874 7,059,598 11,369,087 16,446,322 21,357,400 25,583,233
Total subscriptions 2,754,286 7,069,874 12,792,812 19,881,549 28,230,149 37,890,646 50,237,139
Table 2: Market Share for High-Speed Services (Year-End)

1999 2000 2001 2002 2003 2004 2005 Change in
Market Share
Fiber or powerline 0.113 0.053 0.039 0.028 0.021 0.018 0.009 -0.104
Other wireline 0.221 0.144 0.084 0.061 0.046 0.039 0.017 -0.204
Satellite or wireless 0.018 0.016 0.017 0.014 0.013 0.015 0.076 0.058
DSL 0.134 0.280 0.309 0.326 0.337 0.365 0.388 0.254
Cable 0.513 0.507 0.552 0.572 0.583 0.564 0.509 -0.003
Figure 2: Changing Market Shares

In many cases, the first firms to offer high-speed lines were cable companies selling cable modem
service. They initially acquired a very high market share, but this market share corresponded to a
tiny penetration rate, as not many people subscribed. Cable modem’s 60 percent market share at
the end of 2004 corresponded to a penetration rate of only about 18 percent.31 Thus, even if an
area currently has no broadband service and the government broadband enterprise gets to be the
first mover, it should not expect a high penetration rate over the long term.
Phone companies offering DSL service were usually the second or third market entrants, and they
gradually built a respectable market share.32 Although DSL is available at prices less than or equal
to that of cable modem, the cable companies enjoyed a substantial lead over the phone companies
due to uncertainty over the regulatory status of DSL service. Phone companies had much lower
broadband market shares than the cable companies enjoyed in the early years. By 2004, DSL had
an estimated broadband market share of 37 percent. But since many households do not use
broadband, the penetration rate for DSL is only about 14 percent.33 Analogously, a government
broadband enterprise that becomes the second major competitor in a market should probably
expect a penetration rate more modest than the first mover’s. In 2005, new DSL subscriptions (5.7
million) exceeded new cable modem subscriptions (4.2 million) for the first time.34 This was likely
due to the substantial price reductions offered on DSL in 2005.
Wireless enjoyed substantial growth in 2005. The number of mobile wireless subscribers rose
from 380,000 in June to 3.1 million by December.35 The FCC did not report mobile wireless
subscribership in previous years, but by way of comparison, total subscribers to all wireless and
satellite numbered 550,000 in 2004. Mobile wireless had a market share of 6.2 percent at the end
of 2005.
Other forms of high-speed service have much smaller market shares, corresponding to much
smaller penetration rates. In some cases, such as fixed wireless and satellite, these may be the first
or second competitor in rural areas that lack cable TV service or are too far from phone company
switching offices to utilize DSL. Fiber optics to the home is just beginning to be deployed on a
large scale, and broadband over powerlines has only recently become feasible. The FCC data
suggest that, except in niche applications, alternatives other than cable modem or DSL will have to
offer significant improvements in price or quality before they can gain significant market share.
Verizon’s pricing of its new fiber optic Internet service seems consistent with this point, as current
prices significantly undercut the price of cable modem service at similar speeds.36
In some ways, the data organized by mode of service understate the effects of competition in many
markets. DSL service, for example, can be offered by the incumbent telephone company or by a
competitor who either builds all of its own facilities or leases the incumbent’s phone line and
installs its own switches and transport. As of December 2005, non-incumbents accounted for 3.7
percent of asymmetric DSL lines and 8.2 percent of high-speed connections using telephone
lines.37 Cable modem service is available from incumbent cable operators, but also from
competitive “overbuilders” who have deployed a great deal of fiber and use coaxial cable only to
connect individual homes to the neighborhood fiber node. Numerous studies find that direct cable
10 Reason Foundation
competition leads to lower cable rates.38 Regulators do not systematically gather data on the price
of broadband. A series of six case studies conducted by the U.S. Government Accountability
Office, however, found that high-speed Internet service cost 20-38 percent less in three markets
with competition from new “broadband service providers,” and about the same in the other three.39
By the end of 2005, 67 percent of U.S. zip codes had at least four high-speed Internet providers.
Ninety-three percent had two or more competitors, and only 1 percent had none.40 The lesson for
government systems should be clear: when competing against several established incumbents and
possibly new entrants as well, it could end up with a tiny market share and penetration rate unless
it offers a decisive cost or quality advantage.
There is evidence that municipal cable and Internet services can achieve high penetration rates if
they’re willing to lose a lot of money doing it. And this means taxpayer or ratepayer money. The
communications utility operated by the Cedar Falls, Iowa, utility has video penetration of 47
percent and high-speed data penetration of 37 percent. Video service started in February 1996;
data service started January 1997.41 Lebanon, Ohio’s municipal cable service achieved a
penetration rate of 37 percent in its first year, despite competition from Time Warner. However, it
has always shown substantial operating losses, and monthly rates are subsidized by at least $37 per
household, which suggests the high penetration rate flows from below-cost pricing.42 Similarly,
Ashland Fiber Network in Ashland, Oregon had a 35 percent penetration rate for cable TV and a
40 percent penetration rate for Internet service as of December 2004. However, the system has
posted an operating loss of about $1.5 million each year since 2002. In May 2005, the City
Council voted to subsidize these services with revenues from the city’s electric utility.43 The 2004
operating loss implies a monthly subsidy of $17 for each cable and Internet subscriber, or $34 for a
household that subscribes to both.44
Once customers are won, they can be expensive to keep. Providers of Internet and
wireless services bear significant expenses to deal with “churn”—loss of customers to
competitors.
Once customers are won, they can be expensive to keep. Providers of Internet and wireless
services bear significant expenses to deal with “churn”—loss of customers to competitors. Internet
providers might expect a customer churn rate of 2.5−3 percent per month. That implies a firm can
expect to lose about a quarter of its customers each year!45 In some sense, these costs are
unavoidable. The firm can either try to reduce churn through attractive pricing and promotion to
current customers, or try to replace lost customers with aggressive pricing and marketing to new
customers. Either way, churn imposes a cost that municipalities should recognize when assessing
the feasibility of broadband networks.
After reviewing many cities’ actual experience with cable and broadband enterprises, Balhoff and
Rowe suggest that a municipal fiber enterprise in a larger, competitive market is unlikely to gain
more than 35 percent of residential subscribers, or about 10 percent of households. A fiber

enterprise that is the “first mover” in a smaller community with less competition might do
somewhat better.46 A wireless system might expect to serve about 25 percent of the residential
market and 10-20 percent of the business market.47
In light of this experience, an assumed penetration rate for a municipal system of more than 10
percent in the first year, or 20-50 percent in subsequent years, appears highly unrealistic in most
cases. The only exceptions might be small communities serviced only by expensive alternatives,
or municipalities willing to commit to very large subsidies for their broadband systems.
B. Performance Competition
Dynamic competition is not just about price, and in some cases price may be a much less important
factor than various aspects of quality or performance. Competitive businesses seek to continually
improve performance—or even develop new aspects of performance that were not previously
thought capable of improvement. Wireless phone service provides a case in point. Since the early
1990s, wireless has been less regulated and more competitive than wireline phone service. In
addition to portable phone service, wireless phone companies have offered a continual stream of
new features, including free long-distance service, ringtones, camera phones, video downloads,
connectivity for PDAs, and wireless Internet access.
For government-provided broadband, performance competition means that price is not the only
factor consumers will consider in evaluating whether to take the service. Cheap, or even free,
service may not be a good deal if performance is inferior to alternatives that either exist now or
may develop in the near future. For broadband, performance includes factors like:
How fast is it?

Do speeds slow if more users are on the system?
How safe is my computer from intrusions by other network users?
Can communications be intercepted?
How effective are parental controls or other technologies customers might use to limit access?
Does the presence of parental controls or filtering for other customers inhibit my ability to
access what I want?
Does the system have any features that protect copyrighted material?
How good is the tech support, and what form does it take (phone, e-mail, Web, Internet chat,
24/7)?
Is it wired or wireless?
If wireless, can I receive the signal everywhere I want to use it, in all kinds of weather?
Is the quality good enough to support voice?
Speed is perhaps the most measurable aspect of performance. The Appendix to this study
compares the prices and speeds offered by about 60 government systems with the prices and speeds
offered by many of the leading private DSL, cable modem, satellite, and wireless Internet
providers. The list of 60 government systems is not comprehensive, but it includes all those for
which it was possible to find price and speed data on Web sites or from other readily available
sources in January 2006. In many cases, it appears that the prices and performance of existing
government systems are inferior to those of existing private systems.
The large phone companies make DSL available at prices ranging from $14.95 to $49.99 per
month, depending on the download speed. All of the “majors” sell a download speed of at least 1.5
mb for below $35. Only one government provider can beat this price for a speed of 1.5 mb or
better: Spencer, Iowa, which sells a 6 mb speed for $34.95. Several others (Harlan, Iowa; Easton,
Maryland; Tacoma, Washington; Bristol, Virginia; and Richmond, Indiana) come within a few
dollars for 3 or 5 mb. Only 10 out of 60 government services sell 1 mb or more for less than $35.
For half of these services, the price is within $6 of $35. None beats the offer of 1.5 mb for $14.99
available from SBC. Even if one assumes the phone companies will abandon their current
promotional prices and revert to their 2004 prices in the $30-40 range for 1.5 mb, they are still
attractively priced compared to most of the government systems.
Most private cable modem service costs at least $50-60 per month. Cable modem usually offers
faster download speeds than DSL—typically 3-5 mb minimum, and sometimes as much as 30 mb.
Only half of the government-sponsored cable modem systems offer a speed of 3 mb or higher, and
only eight (25 percent) offer a speed of 5 mb or higher. About one-third of all the government
systems offer a speed of 3 mb or more at a price below $60, which would make them competitive
with the private cable companies. Sixty dollars may, however, may be an artificially high
benchmark, as the phone companies recently made speeds of 3 mb or more available via DSL at
substantially lower prices. Clearly, some government systems offer speeds and prices comparable
to the private systems, but the majority do not.
Another way of quantifying the price/performance relationship is to examine the monthly price per
kilobit of download speed. Except for the relatively slow “entry level” DSL offerings, the phone
companies’ DSL costs 1 or 2 cents per kilobit. Cable modem costs the same or less, and Verizon’s
fiber optic service costs tenths of a cent per kilobit. About one-third of the government services
offer an option that costs less than 2 cents per kilobit. Most of the wireless services cost between 5
and 15 cents per kilobit. On a per-kilobit basis, the private providers have an edge over the
government services.
Of course, one might imagine that a large number of broadband customers merely want something
somewhat faster than dialup. For those consumers, what really matters is the price of the “entry
level” offerings that are several times the speed of dialup. However, the government services
somewhat faster than dialup (128-256 kb) cost about the same as DSL or the “slow” cable modem
option offered by RCN, a competitive broadband service provider. The slower speeds give the
government no price advantage. The principal government services that could be regarded as
exceptions would be the wireless services in Buffalo, MN ($9.99 for 192 kb) and Chaska, MN ($16
for 1 mb), and the fiber service in Kutztown, PA ($15 for 1 mb).
Many of the slower government systems appear to be in rural areas. If these areas lack DSL and
cable, then the relevant competitors would be satellite and, in some cases, wireless. The
government systems are generally less expensive than satellite, and often less expensive than
private wireless. Therefore, they may offer a good alternative in some rural areas.
Many of the current government offerings appear to suffer significant handicaps compared to
private DSL, cable modem, and fiber. Unless the mobility of a government wireless system
provides significant value that most customers are willing to pay extra for, government systems
that currently compete with these technologies will likely need to upgrade their speeds or cut their
prices if they expect to retain customers. New government systems will have to offer higher speeds
or lower prices than most of the existing government systems currently offer if they want to
compete successfully with private DSL, cable modem, and fiber.
An effective government-owned competitor must be prepared to offer a

price/performance combination that a significant number of consumers will prefer to those
offered by competitors.
An effective government-owned competitor must be prepared to offer a price/performance

combination that a significant number of consumers will prefer to those offered by competitors. If
government ignores performance competition, it could end up offering a fairly plain service
appealing only to customers who want relatively slow broadband speeds—and may not be willing
to pay much for it. In effect, government would be seeking an unattractive market niche similar to
the one now occupied by the dialup Internet access firms. While such an approach might be
attractive as social policy, it is unlikely to pay for itself over the long term and would likely require
ongoing subsidies.
This discussion has focused on speed to illustrate the point about performance competition, but
similar reasoning applies to other aspects of broadband performance that customers think are
important. A successful competitor need not be better than the competition on all quality
attributes. At a minimum, though, a government broadband enterprise needs to carefully consider
how the major quality attributes of its service compare to the alternatives that are either available in
the market today or are likely to materialize in the future.
C. Continuous Improvement
Dynamism means change. In a dynamic market, business must continually improve in order to
retain customers. Firms that fail to stay abreast of technological improvements tend to lose ground
to those that do.
Price Improvement
One indicator of the extent of change is the pace at which prices fall as technology improves, costs
fall, or competition intensifies. This has occurred frequently in the market for Internet service, as
well as in related or analogous markets such as wireless communications, telephone equipment,
and telecommunications services. As Figure 3 shows, the real consumer price index for Internet
services has fallen by 23 percent since the Bureau of Labor Statistics started tracking this series in
1997. Real consumer price indices for wireless, telephone equipment, and long-distance service
have fallen even faster—by 45-65 percent.48
Figure 3: Real Consumer Price Index, Internet Services
130.00
125.00
120.00
115.00
110.00
105.00
100.00
95.00
90.00
1997 1998 1999 2000 2001 2002 2003 2004 2005
Source: Bureau of Labor Statistics
A more direct measure of the speed of progress in broadband can be found in substantial price
reductions that have occurred in recent years. Between 2004 and 2005, Bellsouth cut the monthly
price of 1.5 mb DSL from $39.95 to $32.95, a 17 percent drop. Qwest dropped its promotional
price from $26.99 to $19.99, and extended the term from three months to a year. SBC cut its
promotional price, good for a year, from $26.95 to $14.95.49 Verizon Wireless reduced the
monthly fee for wireless broadband service using a PC card by 25 percent, from $79.99 to
$59.99.50
Performance Improvement
Continuous improvement may also be important in performance as well as price. One indicator
that performance competition matters in broadband is the rate at which maximum speeds have
increased. In its first report on the extent of broadband deployment, the FCC chose to define “high
speed” service as Internet service capable of at least 200 kbps one-way. “Advanced services” lines
are defined as connections capable of at least 200 kbps both ways. The FCC has continued to
employ those definitions to ensure data comparability in subsequent years.
Figure 4: Monthly Consumer Cost of Internet Access, 1998
$100.00
Amortized equipment Amortized installation Monthly fees
$90.00
$80.00
$70.00
$60.00
$50.00
$40.00
$30.00
$20.00
$10.00
$0.00
Dialup ISDN Satellite Terr. Terr. ADSL Cable
wireless -- wireless -- modem
MMDS LMDS
Source Author’s calculations based on data in FCC, Broadband Report 1999, CC Docket No. 98-146 (January 28, 1999).
Many forms of broadband service available today, however, are much faster than that, and speeds
have improved rapidly. Between 2004 and 2005, a number of major broadband providers
increased the speed of their service. SBC increased the upload speed of its DSL service threefold,
from 128k to 384k. Cablevision increased its download speed from 5 mb to as much as 10 mb.
Comcast increased its download speed from 3 mb to 4 mb and its upload speed from 256k to 384k.
Time Warner increased download speed from 3 mb to as much as 8 mb.51 These changes represent
performance improvements of between 25 percent and 200 percent—in one year. In January 2006,
Web pages indicated further improvement in maximum speeds. Comcast offered a maximum
download speed of 6 mb, Cox offered 15 mb, and Cablevision offered 30 mb.52
Figure 5: Residential Internet Speeds Available in 1998 (kilobytes per second)

3500
3000
2500
2000
1500
1000
500
0
Dialup ISDN Satellite Terr. Terr. DSL Cable modem
Wireless Wireless
MMDS LMDS
Source: FCC, Broadband Report 1999, CC Docket No. 98-146 (January 28, 1999).
If recent experience is a guide, government broadband operations will need to be prepared to

continually improve in the future if they want to keep pace with private sector competitors.
Government broadband operations will need to be prepared to continually improve in the

future if they want to keep pace with private sector competitors.
D. Technological Change and Lock-In
A vigorous scholarly debate rages over whether private firms make avoidable choices that “lock
in” inferior technologies. “Lock-in” occurs when an initial decision gives one technology a slight
edge, then sets in motion a process which leads that technology to dominate the market. If the
technology that gets locked in is truly the superior technology, then there’s no harm done. But if
an inferior technology gains a temporary edge in market share, some scholars argue that it might
remain dominant even though it is inferior. The head start could allow an inferior technology to
dominate because costs fall as the number of users increases (“increasing returns”) or because the
value of the technology to each user increases as more people use it (“network effects.”)53
Discussions of lock-in often focus on numerical examples in which the early adopters of
technologies choose the technology that creates the biggest payoff in the short term. An alternative
technology, however, creates greater value over the longer term, when there is a larger number of
adopters. The market gets locked in to the inferior technology due to the decisions of the early
adopters. Lock-in arguments thus assume that the first decisionmakers cannot find a way to profit
from forgoing the immediate short-term benefits of selecting the inferior technology so that they
can reap the larger, longer-term benefits of the superior technology.54
Such lock-in may or may not be plausible in private markets. However, it is quite possible in
government enterprises, because a wide variety of barriers reduce decisionmakers’ ability and
incentives to select the technology that is superior over the long term. Even those who are
skeptical about lock-in allegations in private market settings admit that public policy can involve
various forms of lock-in. Potential sources of sub-optimal initial decisions include the influence of
concentrated interest groups, short political time horizons due to election cycles, less intense
competitive pressures because voters must select from among candidates offering bundles of
positions on different issues, and limitations on government decisionmakers’ ability to profit
personally from making decisions that most benefit the public.55
If policymakers choose the “wrong” technology, however, increasing returns or network

effects may prevent or delay adoption of better or less expensive technologies that might
otherwise come sooner.
Public policies run a significant risk of locking in technologies that quickly become outdated. In
the mid-1990s, for example, public utility commissions and consumer groups sought mandates to
ensure that telephone companies would hasten the rollout of the only “high-speed” Internet
technology believed to be feasible anytime soon: ISDN. “As it turned out,” note Scott Wallstein,
“ISDN was a fairly short-lived technology that was expensive and relatively slow, supporting
speeds only up to 128 Kbps, or about three times as fast as a dialup modem.”56 By 1998, survey
data gathered by the FCC revealed that ISDN was the most expensive and second-slowest form of
Internet access. Broadband subscriptions took off shortly thereafter, rising from 1.8 million in
1999 to 50 million in 2005.57
Governments that want to get into the broadband business already face several technology choices.
Municipal cable operators can easily offer cable modem service. For new builds, the two most
commonly discussed options are fiber optic and Wi-Fi. If policymakers choose the “wrong”
technology, however, increasing returns or network effects may prevent or delay adoption of better
or less expensive technologies that might otherwise come sooner. Of course, it is difficult to
predict in advance what some of those advances might be, but several plausible options include:
Further advancements to boost speed and reduce congestion in cable modem service
Further advancements to boost speed of DSL service
Full or partial deployment of fiber optic service by cable or phone companies
“Evolution Data Optimized” technology that allows wireless phone companies to offer
wireless broadband at speeds of 400-800 kb.58
Broadband over powerlines
Improvements to reduce the cost of satellite broadband
WiMAX, a more powerful wireless option
Subsidies could exacerbate lock-in. Public subsidies for broadband could be funded by taxes or by
excessive charges on monopolized services provided by the government enterprise. In addition, a
government broadband enterprise could receive an implicit subsidy in the form of costless, below-
cost, or perhaps even exclusive access to the public rights-of-way.
If subsidies allow a government enterprise to offer broadband service at a price that fails to cover
costs, then competitors face a higher bar to successful market entry, even if they have a better
technology. Suppose, for example, the government offers 200 kilobyte Internet access for $10 per
month, even though it costs $20 per month to produce. Suppose further that private competitors
could offer 10 megabyte service for $40 per month. Many consumers might prefer the faster
service at $40 to the slower service at $20, but they’ll choose the slower service if it only costs
$10. If the government service is subsidized, the competitor cannot afford to introduce its faster
service until further technological progress either improves the quality or reduces the cost
sufficiently to let it attract consumers away from the subsidized service. Until that happens,
consumers have to content themselves with the slower, subsidized service.
The point here is not just that lock-in via subsidies wastes the public’s money, but also that
consumers have to wait longer to get a better service, because competitors are deterred by the
subsidy. Consumers would be better off if the price of the government service were not
subsidized, because competitors would provide the superior combination of service and price
sooner.
Government broadband plans should squarely address the potential for lock-in and explicitly
evaluate whether subsidies would give an inferior technology an artificial boost.
E. Obsolescence
Monopolized government enterprises (and traditional regulated, investor-owned utilities as well)

can afford to stretch out the recovery of their capital costs over a long period of time. Their prices
are lower as a result. There are two reasons such enterprises can use long depreciation schedules:
they are often monopolies, and technological progress is relatively slow in any event.
In a dynamically competitive market, networks become obsolete faster. Technology improves

more rapidly, and competitors are not shy about introducing the newest technology if they feel it
will give them a cost or quality edge over the competition. As a result, capital investment becomes
obsolete more quickly. Accurate cost recovery requires higher prices in the early years than under
static monopoly, so that the firm can recover its capital costs before competitors’ improvements
constrain its ability to do so.59
If capital becomes obsolete more rapidly, the firm must recover its capital costs over a shorter
period of time. Business plans for government broadband enterprises need to assume faster
depreciation rates, and concomitantly higher prices, than have traditionally been used for
government utilities. Balhoff and Rowe, for example, argue that Wi-Fi will have to be depreciated
over 3–5 years, versus 15 or more years for fiber.60 Thus, a workable plan for municipal Wi-Fi
needs to assume that revenues will not just cover operating costs plus interest, but also recover the
initial capital outlay in 3–5 years.
F. Risk
Obsolescence is not just faster under dynamic competition; it is also less predictable. Financially,
the investment is less of a “sure thing” than a conventional government monopoly. That means the
cost of capital should carry a higher risk premium than normally considered appropriate for
government enterprises. A higher cost of capital, in turn, implies that the price consumers pay
should be higher if the enterprise is expected to cover its costs.
Are broadband enterprises riskier than other infrastructure investments that governments
sometimes make? Assessing the inherent riskiness of government enterprises is notoriously
difficult. Even assessments based on financial market evaluations (such as interest rates on bonds)
may be misleading, because investors will regard explicit or implicit government guarantees or
subsidies as factors that reduce risk. The underlying business might face substantial risks, yet
financial markets might assign very little risk premium to bonds because guarantees or subsidies
transfer the risks from investors to taxpayers or ratepayers of affiliated government enterprises.
However, one can gain some idea of the relative riskiness of various types of businesses by
examining how financial markets evaluate the riskiness of private firms in different industry
sectors. A common measure of risk is the “beta coefficient” calculated by financial analysts. The
beta coefficient measures how risky a stock or industry sector is compared to the stock market as a
whole. An industry with a beta of 1, for example, is neither more nor less risky than the overall
stock market. A beta below 1 signifies relatively less risk than the market as a whole, and a beta
exceeding 1 implies relatively more risk than the market as a whole.
Table 3 reports recent beta estimates for various industry sectors that are similar to different types
of government infrastructure enterprises. All betas were calculated in January 2005 based on the
previous five years of stock market data.61
Table 3: Utility Risk Levels

Low Risk Average Beta High Risk Average Beta
Electric Utility 0.72-0.79 Cable TV 1.75
Gas Distribution 0.65 Telecom Services 1.32
Water Utility 0.60 Wireless Networking 2.38
Source: Aswath Damodaran, http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/Betas.html
It is clear from this table that investments in electric, gas, and water utilities have involved much
less risk than investments in firms that sell broadband or wireless data services. Electric, gas, and
water are precisely the types of static, monopolized industries where governments have
traditionally invested. In terms of risk, broadband is a whole new ballgame. Investing is
broadband is much riskier than investing in the overall stock market.
Nevertheless, some governments have financed broadband initiatives as if they were traditional
low-risk investments in infrastructure that provides necessities:
Spencer, Iowa’s Spencer Municipal Communications Utility has an $8 million loan from the
city’s electric utility at 4.5 percent, whereas the electric utility’s revenue bonds have an interest
rate of 5.75 percent.62 This rate differential implies that the communications utility is less risky
than the electric utility! The municipal utility also allocated $11.7 million worth of
communications infrastructure costs to the electric utility, effectively financing them at 5.75
percent.63 The utility offers cable, high-speed data, and telephone service in competition with
Mediacom and Qwest.
Cedar Falls Communications Utility pays 5 percent interest on a $6 million loan from an
affiliated city utility.64 It offers cable and high-speed data. Cedar Falls is regarded as one of
the best-run municipal fiber operations with some of the best financial results.65
Lebanon, Ohio financed construction of its hybrid-fiber-coax network with $5 million of 4-
year revenue bonds at 5.26 percent, a $1 million 1-year note at 2.7 percent, and a $750,000
one-year note at 3 percent.66
Bristol, Virginia financed its fiber optic system with a combination of $15 million in loans
from the city electric utility at rates ranging from 3.77-5.99 percent; revenue bonds at rates
between 2.0 and 5.35 percent; and a $2 million grant from the Virginia Tobacco
Indemnification and Community Revitalization Commission.67
The 14-city UTOPIA fiber project in Utah assumes it can finance the project at 6 percent over
20 years.68
Figure 6: Utility Risk Levels

(Beta Coefficients, Where Stock Market = 1)
2.5
1.5
0.5
0
Electic Utility Gas Distribution Water Utility Cable TV Telecom Wireless
Services Networking
Data Source: Aswath Damodaran, http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/Betas.html
Many of these interest rates seem low when compared with analysts’ estimates of the “true,” risk-
adjusted cost of financing government broadband initiatives. Balhoff and Rowe, for example,
suggest that the true cost of capital for the Lafayette fiber project is likely 10-13 percent, given the
risk involved.69
Another implication of high risk is that it takes a different kind of organization, with different
strategies, to deal with the risk, especially when future events are difficult to predict. As Aaron
Wildavsky pointed out, high risk makes a strategy of “resilience” more appropriate than a strategy
of “anticipation.” When technological risks are low, well-understood, and have known remedies,
then decisionmakers can best deal with them by anticipating what will happen, taking preventive
measures, and developing plans to cover every contingency. Expert knowledge and the ability to
plan are key resources needed by organizations in this type of environment. When risks are large
and remedies unknown, then organizations need to develop “resilience:”
A strategy of resilience … requires reliance on experience with adverse consequences once
they occur in order to develop a capacity to learn from the harm and bounce back. Resilience,
therefore, requires the accumulation of large amounts of generalizable resources, such as
organizational capacity, knowledge, wealth, energy, and communication, that can be used to
craft solutions to problems that the people involved did not know would occur. Thus, a
strategy of resilience requires much less predictive capacity but much more growth, not only in
wealth but also in knowledge. Hence it is not surprising that systems, like capitalism, based on
incessant and decentralized trial and error accumulate the most resources.70
An accurate cost of capital provides important discipline that is especially critical in a high-
risk environment.
A traditional municipal utility is more likely to face a low-risk environment in which anticipation
is a viable strategy. A broadband enterprise faces a high-risk environment that demands resilience.
The cost of capital helps a private firm determine how much it should invest in developing the
capabilities that produce resilience, and what kinds of investments to make. A government
enterprise that faces an artificially low cost of capital is more likely to waste the public’s money by
“investing” in capabilities that produce little value for customers, or do so only after an excessively
long time. An accurate cost of capital provides important discipline that is especially critical in a
high-risk environment.
G. Uncertainty
The term “risk” describes a situation in which the actual outcome in the future is unknown, but it is
possible to quantify or estimate the possible outcomes and the probabilities that they will occur. In
addition to risk, firms in dynamic markets face uncertainty—the possibility of unique but
significant events that cannot easily be quantified.71
Development of WiMAX technology may be one example of uncertainty faced by government

broadband projects. WiMAX seems to hold tremendous promise but has not yet been deployed
anywhere in the United States. Indeed, the Institute of Electrical and Electronics Engineers has not
yet settled on a WiMAX standard.72
Another significant uncertainty comes from wireless phone companies, who are launching wireless
Internet service at speeds of 400-800 kb. This service uses a technology called Evolutionary Data
Optimized, or EV-DO. Although one major wireless company markets this service as an
alternative to Wi-Fi, at this point it is simply not clear whether such products are substitutes for or
complements to Wi-Fi.73
The market potential of both WiMAX and EV-DO is unclear, yet both could be powerful
competitors to Wi-Fi—perhaps even making Wi-Fi obsolete. The likely outcomes and
probabilities may not be well-enough understood to consider these technologies a risk to Wi-Fi;
they are more likely sources of uncertainty.
A private business firm’s shareholders bear uncertainty as well as risk. The prospect of additional,
higher returns entices them to bear that uncertainty. The fact that uncertainty affects shareholders’
financial fortunes gives them strong incentives to seek out management that will exercise sound
judgment.74
How to replicate these incentives, so that decisionmakers in government enterprises make sound
judgments in the face of uncertainty, is a conundrum. The most likely method would be to
organize the enterprise as a for-profit company, with explicit expectations from the owner (the
government) that it be commercially successful. The government would have to commit to
allowing the enterprise to operate in a commercial fashion—including, for example, permitting
hiring and compensation policies similar to those found in the private sector. The most credible
way governments make these types of commitments is by enacting a plan to privatize the
enterprise. But in this context, a privatization plan would beg the question of why the government
is getting into the broadband business to begin with! Thus, for broadband, government faces the
fascinating challenge of committing to commercial operation without promising privatization.
For government broadband enterprises, taxpayers bear the uncertainty in their role as the ultimate
owners. At a minimum, therefore, effective accountability requires that government broadband
initiatives should have accountability and transparency for taxpayers at least as good as that which
publicly held companies must have for their shareholders. Basic criteria for accountability include:
Publicly available, audited financial statements prepared according to Generally Accepted

Accounting Principles
Transparent disclosure and quantification of any subsidies
Clear articulation of public benefits that the government enterprise is supposed to produce,
along with an assessment of the evidence that these benefits will occur
Outcome-oriented performance measures that will reveal whether the government enterprise
causes these benefits to occur
Data on performance and public benefits that are as verifiable and valid as the financial data
These transparency measures may not be sufficient to make government managers as accountable
to uncertainty-bearing taxpayers as corporate managers are to uncertainty-bearing owners. But it is
difficult to see how accountability is possible without them.
Part 4
“Free” Networks: Beware of Geeks

Bearing Gifts
M ost of this study has focused on broadband initiatives that actually require some type of
public sponsorship or investment—either by government or by government-owned entities.
A recent twist, which appears to offer the public a much better deal, has arisen in several cities,
where private firms have proposed to build Wi-Fi networks at no cost to taxpayers. Cities where
such proposals have been mentioned include Philadephia, San Francisco, Milwaukee, Houston, and
Mountain View, California, the home of Google and other high-tech firms.75
In Philadelphia, Earthlink won a high-profile contract to build a Wi-Fi network covering 135
square miles by placing transmitters on telephone poles. CNET News noted, “The City of
Brotherly Love's plans differ from those of many other municipalities in one crucial way:
EarthLink will own the hardware and take the financial risk associated with providing the service.
If it flops, city taxpayers won't lose the money.”76 The goal (not guaranteed) is to offer Wi-Fi
service for $20 per month, or $10 for low-income users.77
San Francisco solicited bids from private sector firms seeking to offer Wi-Fi service using the
city’s right-of-way and light poles. Google proposed to build a citywide Wi-Fi system at no charge
to the city, plus offer “free,” advertiser-supported Wi-Fi service at a speed of 300 kilobits per
second.78
Houston Mayor Bill White announced in November 2005 that his city wants a piece of this action.
Existing broadband competitors, he contends, have “no good reason to fight it unless we were
going to try to finance the system with taxpayer dollars, which we're not proposing to do. We're
not trying to compete with any private business, and we want to make this technology option
available.”79
It is unlikely that such firms are offering “free” (or below-cost) Wi-Fi out of charitable motives, or
merely to build goodwill. Presumably they seek to offer Wi-Fi because they see a revenue stream
somewhere, even if the users or the local government are not charged directly for Wi-Fi. This is
just a larger-scale version of the business model followed by some hotels and retailers, who offer
“free” Wi-Fi because they believe it will bring in more paying customers. Google’s “free” Wi-Fi
would be supported by advertising, and perhaps other revenue streams that Google expects will
increase as more people use its Wi-Fi service. As Google noted in its Mountain View proposal,
“Google is in a unique position to offer free access to the Internet, since we have the ability to
subsidize and earn revenue indirectly when these Wi-Fi users get on the Internet and access our
and our partner sites.”80
There is nothing inherently wrong with proposals for free or privately subsidized Wi-Fi. However,
governments need to realize that rights-of-way and light poles are valuable assets, and access to
these assets would bestow a significant competitive advantage on any firm selected to use them.
Utility regulators have noted that access to rights-of-way is one of the most significant obstacles to
broadband deployment. The California Public Utilities Commission noted in 2005, “The process
for obtaining Right of Way permits for construction of broadband infrastructure in California is
lengthy, expensive, inconsistent and is cited as one of the most significant barriers to broadband
deployment.”81 An AEI-Brookings study by Scott Wallstein finds that states with laws that
guarantee telecommunications companies access to right-of-way have broadband penetration that
is about 10 percent higher than states which do not guarantee access. States that fail to extend this
guarantee to cable companies have somewhat lower broadband penetration.82
Indeed, the ongoing Wi-Fi “gold rush” resembles the cable TV gold rush of the 1970s. Once the
FCC stopped protecting broadcasting from cable competition, the race was on to wire America’s
homes. In most locations, cable companies competed for monopoly franchises by offering cities a
variety of “free” or below-cost services, such as channels for public, educational, and government
access; studios for creation of public access programming; and wiring of various public facilities.
Consumers ultimately paid for these freebies via higher cable rates, which were only possible
because local governments typically permitted only one cable company to use the public rights-of-
way. Two decades of economic research have amply demonstrated that cable rates are
significantly lower when more than one cable company competes for consumers’ business.83
Largely because of this evidence, the federal government now prohibits local governments from
granting monopoly cable franchises or “unreasonably” refusing to grant a competitive franchise.84
Local governments should beware of granting one Wi-Fi provider exclusive access to
public assets, even if the Wi-Fi service itself is free of charge to users.
In the Wi-Fi gold rush, government or private enterprises receive exclusive use of light poles and
other city property for placement of transmitters. In exchange, the Wi-Fi provider often offers
some kind of bargain service for local government and a free or low-cost entry-level service for
citizens. Where the money comes from to pay for the Wi-Fi freebies remains to be seen.
Any local government that grants one Wi-Fi provider an exclusive right to use right-of-way and
poles risks distorting competition in whatever markets are generating the revenue stream that will
subsidize the Wi-Fi service. The only provider of advertiser-supported Wi-Fi allowed to use city
light poles, for example, could likely charge a higher price for advertising than if there were
competing providers of advertiser-supported Wi-Fi. A monopoly that gives away Wi-Fi to build
demand for other services it might sell to Wi-Fi users might be able to charge a higher price for
these other services than it would in the presence of other Wi-Fi competitors.
For this reason, local governments should beware of granting one Wi-Fi provider exclusive access
to public assets, even if the Wi-Fi service itself is free of charge to users. At a minimum,
decisionmakers should assess whether exclusive access would distort competition in the markets
for other goods and services sold by the Wi-Fi company. They should seek the assistance of the
Federal Trade Commission or the Antitrust Division if they lack the capabilities to perform such an
assessment.
This warning should not be controversial, as it is consistent with public policy on wired broadband
provided by for-profit companies. Nowadays no one would seriously advocate that only the cable
company or only the phone company should be permitted to use the public right-of-way to provide
broadband service. Similarly, the Telecommunications Act of 1996 explicitly prohibits state and
local governments from granting exclusive franchises for telecommunications service. National
policy explicitly envisions competition in broadband, and the Federal Communications
Commission is required by law to report annually on the state of broadband deployment and
competition.85 A 2006 Federal Trade Commission staff report likewise recommends that local
governments providing Wi-Fi should ensure that they do not disadvantage or exclude private
competitors.86 Local governments should not let the sizzle of free Wi-Fi obscure the consumer’s
stake in competition.
Part 5
Conclusion
T he factors outlined above need not imply that government-provided broadband is a bad idea.
However, no plan for government-sponsored broadband should be considered complete or
responsible unless it addresses the factors outlined in this report. More specifically, any plan that
adequately takes dynamic competition into account should answer the following questions:
Who are the competitors?

Do the projected penetration rate and market share reflect realistic assumptions about current
and potential competition?
What cost or quality advantage will the government broadband service have over competitors?
What performance attributes matter to customers?
How will the government service compare to competitors on these performance attributes?
How rapidly have broadband prices fallen in the market where the government enterprise will
compete?
How rapidly can prices be expected to fall in the future?
How rapidly have performance attributes improved in the market where the government
enterprise will compete?
How rapidly can performance be expected to improve in the future?
How will the government enterprise keep pace with competitors’ price reductions and/or
performance improvements?
What safeguards are necessary to ensure that the government enterprise competes with the
private sector strictly on the merits?
What safeguards are necessary to ensure that any benefits the government enterprise receives
due to its public nature do not distort competition in related markets?
What safeguards are necessary to ensure that private broadband competitors obtain access to
public rights-of-way on equal terms with the government entity?
What safeguards are necessary to ensure that political considerations or subsidies do not let an
inferior technology dominate the market by giving it a head start?
What is the economically useful life of the capital deployed by the government enterprise and
the appropriate depreciation rate?
What cost of capital reflects a realistic depreciation rate on the assets?
What cost of capital accurately reflects the risks that the government enterprise will face in a
dynamically competitive market?
How will the incentives of the enterprise and its employees be structured to foster sound
judgment in the face of significant uncertainty about competition and future technological
developments?
How will any subsidies be quantified and transparently disclosed to the public?
What public benefits is the enterprise intended to produce?
How will these benefits be measured and communicated to the public?
How will financial, performance, and public benefit data be gathered, verified and validated?
These questions may appear daunting. They are no less daunting than the challenge of actually
entering a market where technological change is swift, the future is uncertain, and competitors’
actions are unpredictable. Such is the nature of dynamic competition. If the questions seem
unfamiliar to policymakers, that’s one more bit of evidence that the nature of dynamic competition
is fundamentally different from the stable, predictable utility markets that have traditionally
attracted public investment.
Appendix A
Private and Public Broadband Providers
Government Systems Name Monthly Fee Download Speed Price/

Location (kilobits/sec.) Kilobit
Wireless
Cupertino, CA MetroFi $19.95 1000 $0.020
Rochelle, IL Rochelle Muni. Utils. $74.95 256 $0.293
$94.95 512 $0.185
Richmond, IN Richmond Pwr. & Lt. $39.95 1000 $0.040
Scottsburg, IN Scottsburg C3bb $35.00 512 $0.068
$70.00 1000 $0.070
Tell City, IN Tell City Elec. Dept. $29.95 384 $0.078
$44.95 768 $0.059
$84.95 1540 $0.055
Western Kansas Wheatland Electric $37.00 512 $0.072
$87.00 1000 $0.087
Owensboro, KY Owensboro Muni. Utils. $29.99 512 $0.059
Vivian, LA, and Fastline Internet $10.00 64 $0.156
Linden, TX $60.00 1000 $0.060
Alexandria, MN Alex. Bd. of Pub. Wks. $29.95 128 $0.234
$39.95 512 $0.078
Buffalo, MN Buffalo Muni. Util. $9.99 192 $0.052
Chaska, MN City-owned ISP $16.00 1000 $0.016
Grand Haven, MI Ottawa Wireless $15.00 100 $0.150
$45.00 512 $0.088
Carthage, MO Ecarthage.com $39.95 1000 $0.040
Marshall, MO Marshall Muni. Utils. $30.00 250 $0.120
$70.00 500 $0.140
$105.00 750 $0.140
Rio Rancho, NM Azulstar $20.00 256 $0.078
$40.00 1500 $0.027
$80.00 4000 $0.020
Floresville, TX Floresville Elec. Lt. & $49.95 128 $0.390
Power $59.95 256 $0.234
$69.95 384 $0.182
$89.95 512 $0.176

$105.95 768 $0.138
$155.95 1024 $0.152
$199.95 1536 $0.130
Benton County, WA Maverick Wireless $19.95 128 $0.156
$34.95 512 $0.068
$49.95 1000 $0.050
Southeast WA Columbia Rural Electric $40.00 256 $0.156
$260.00 1500 $0.173
Sun Prairie, WI Sun Prairie Wtr./Lt. $35.00 768 $0.046
Cable Modem
Opp, AL Opp Cablevision $24.95 256 $0.097
$34.95 512 $0.068
$44.95 1024 $0.044
Scottsboro, AL Scottsboro Elec Pwr Bd $31.00 512 $0.061
$43.00 1500 $0.029
$58.00 3000 $0.019
Ketchikan, AK Ketchikan Pub Util $47.95 512 $0.094
$59.95 1000 $0.060
$99.95 1500 $0.067
Conway, AR Conway Corp. $39.95 2048 $0.020
$59.95 3084 $0.019
Paragould, AR Parag. Lt., Wtr. & Cable $25.95 NA
$39.95 NA
$59.95 NA
Alameda, CA Alameda Pwr & Teleco $29.99 1000 $0.030
$49.99 3000 $0.017
$52.99 4000 $0.013
Elberton, GA City $50.00 500 $0.100
Monroe, GA Monroe Utilities $39.95 6000 $0.007
Algona, IA Algona Muni. Utils. $49.95 1000 $0.050
$69.95 1500 $0.047
Alta, IA City of Alta $44.95 256 $0.176
$54.95 512 $0.107
Cedar Falls, IA Cedar Falls Utilities $24.95 168 $0.149
$40.00 3920 $0.010
Harlan, IA Harlan Muni. Utils. $37.50 1540 $0.024
Laurens, IA Laurens Muni. Comm. $49.95 1000 $0.050
Muscatine, IA Muscatine Power $21.95 128 $0.171
$39.00 1000 $0.039
$59.95 3000 $0.020
Orange City, IA Orange City Communic. $39.95 3000 $0.013
Osage, IA Osage Muni. Utils. $45.95 256 $0.179
$59.95 512 $0.117

$79.95 768 $0.104
$99.95 1000 $0.100
Spencer, IA Spencer Muni. Utils. $34.95 6000 $0.006
$64.95 10000 $0.006
Sanborn, IA The Community Agency $26.95 128 $0.211
$39.95 512 $0.078
$69.95 1000 $0.070
Bardstown, KY Bardstown Cable TV $27.95 512 $0.055
$32.95 1000 $0.033
$42.95 3000 $0.014
$52.95 6000 $0.009
Frankfort, KY Frankfort Elec. & Water $19.00 128 $0.148
$23.00 256 $0.090
$29.00 512 $0.057
Glasgow, KY Glasgow Elec. Board $25.95 1000 $0.026
Murray, KY Murray Electric $29.95 256 $0.117
$40.95 500 $0.082
$48.95 1000 $0.049
$59.95 4000 $0.015
Easton, MD Easton Utils. Comm. $24.95 128 $0.195
$39.95 5000 $0.008
Braintree, MA Braintree Elec. Light $39.00 5000 $0.008
Poplar Bluff, MO City of Poplar Bluff $29.95 256 $0.117
$34.95 512 $0.068
Lebanon, OH City/GO Concepts $45.95 5000 $0.009
$74.95 10000 $0.007
Wadsworth, OH Elec. & Comm. Dept. $22.45 128 $0.175
$29.95 256 $0.117
Brookings, SD Brookings Muni. Utils. $34.95 512 $0.068
Columbia, TN Columbia Pwr. & Water $32.95 384 $0.086
$41.95 640 $0.066
$47.95 1500 $0.032
$57.95 2000 $0.029
$77.95 3000 $0.026
Fayetteville, TN Fayetteville Electric $44.95 256 $0.176
Greenville, TX Greenville Elec. Util. $37.95 6000 $0.006
Tacoma, WA Click/Advanced Stream $29.90 1000 $0.030
$39.90 3000 $0.013
$59.90 6000 $0.010
Fiber
Sylacauga, AL Sylacauga Util. Board $120.00 1000 $0.120
Ashland, OR Ashland Fiber Network $44.00 5000 $0.009

Bristol, VA Bristol Virginia Utilities $26.36 1000 $0.026
$35.16 3000 $0.012
$39.56 5000 $0.008
Sallisaw, OK Sallisaw DiamondNet $29.95 1000 $0.030
$39.95 2000 $0.020
$59.95 4000 $0.015
$149.95 10000 $0.015
Kutztown, PA Hometown Utilicom $15.00 1000 $0.015
$20.00 1000 $0.020
$25.00 1000 $0.025
$30.00 1000 $0.030
$40.00 1000 $0.040
Reedsburg, WI Reedsburg Util. Comm. $24.95 128 $0.195
$34.95 1000 $0.035
$39.95 3000 $0.013
DSL/ISDN
Gainesville, FL (ISDN) GRUCom $24.95 128 $0.195
Richmond, IN Rich. Pwr. & Lt. $39.95 3000 $0.013
Barnesville, MN Barnesville Muni. Tel. $41.90 128 $0.327
$48.85 256 $0.191
$57.85 384 $0.151
$74.85 512 $0.146
Appendix B
Notes on Sources and Calculations

Most of the list of government Internet services was compiled by Steve Titch and the author from
two sources:
1. American Public Power Association, 2004-05 Annual Directory and Statistical Report.
2. FTTH Council, Fiber Optic Communities of the United States, and Telecommunications
Industry Association, U.S. Optical Fiber Communities 2005 (May 2005).
A few additional providers were not listed in these directories but have been mentioned in press
reports or in Balhoff and Rowe (p. 27).
Price and speed data were acquired from the providers’ Web pages in January 2006. Providers that
did not post prices and speeds on their web pages are not included in this table. Data for some
additional government wireless services are from table in Balhoff and Rowe (p. 27).
Data for private Internet services were acquired from the providers’ Web pages in January 2006.
Prices are for unbundled high-speed Internet service and include charges for modem rental, where
such information is available. Prices omit the effects of one to three month promotional pricing,
but may include offers whose prices are good for one year.
About the Author
Jsince
erry Ellig has been a senior research fellow at the Mercatus Center at George Mason University
1996. Between August 2001 and August 2003, he served as deputy director and acting
director of the Office of Policy Planning at the Federal Trade Commission while on a leave of
absence from the Mercatus Center. Dr. Ellig has also served as a senior economist for the Joint
Economic Committee of the U.S. Congress and as an assistant professor of economics at George
Mason University.
Dr. Ellig has published numerous articles on government regulation and business management in
both scholarly and popular periodicals, including the Journal of Regulatory Economics,
Managerial and Decision Economics, Business & Politics, Antitrust Bulletin, Contemporary Policy
Issues, Competitive Intelligence Review, Journal of Private Enterprise, Texas Review of Law &
Politics, The Wall Street Journal, New York Times, Barron’s, and the Washington Post. His co-
authored/edited books include Dynamic Competition and Public Policy (Cambridge, 2001),
Economic Deregulation and Customer Choice (Center for Market Processes, 1997, with Robert W.
Crandall), New Horizons in Natural Gas Deregulation (Praeger, 1996, with Joseph Kalt) and
Municipal Entrepreneurship and Energy Policy (Gordon & Breach, 1994, with Alison E.
Woodward and Tom R. Burns).
Dr. Ellig received his Ph.D. and M.A. in economics from George Mason University in Fairfax,
Virginia, and his B.A. in economics from Xavier University in Cincinnati, Ohio.
Related Reason Studies

Spinning its Wheels: What Other Cities Can Learn from iProvo’s First 18 Months of Municipal
Broadband, by Steven Titch, Reason Foundation Policy Study 348, August 2006,
http://www.reason.org/ps348.pdf
The Municipal Broadband Compact, September 2005,

http://www.reason.org/wifibroadband/municipalbroadbandcompact.shtml
Questions Public Officials Should Ask About Government Broadband Services

September 2005, http://www.reason.org/wifibroadband/broadbandquestions.shtml
Spectrum Privatization: Removing the Barriers to Telecommunications Competition, by David

Colton, Reason Foundation Policy Study No. 208, July 1996, http://www.reason.org/ps208.html
On the Frontier of Deregulation: New Zealand Telecommunications & the Problem of

Interconnecting Competing Networks, by Milton Mueller, Reason Foundation Policy Study No.
177, May 1994, http://www.reason.org/ps177.html
Endnotes
1
“Survey Finds 1 in 5 Americans Have Never Used the Web,” Barre, VT, Times Argus, October
9, 2005.
2
Michael J. Balhoff and Robert C. Rowe, Municipal Broadband: Digging Beneath the Surface,
Balhoff & Rowe, LLC September 2005, p. 25.
3
“Broadband Customer Rise Aids Telecom,” Des Moines Register November 11, 2005.
4
Balhoff and Rowe, Municipal Broadband, p. 16.
5
“A Tale of Two Cities,” Information Week, October 10, 2005; “Phone on the Range,”
Entrepreneur October 2005.
6
Eric Auchard, “S.F. Mayor Sees Wireless Service as a Basic Right,” Reuters, October 3, 2005.
7
“Area’s Wi-Fi Plans Slowly Progressing,” Peoria Journal Star, March 21, 2006.
8
“Manassas Ready to Model BPL Success,” Internetnews.com, October 6, 2005.
9
For more information on WiMAX, see http://www.wimaxforum.org/about/faq.
10
Of course, local officials may also choose either to subsidize infrastructure or to employ
excessive user charges as a form of hidden taxation.
11
Joseph A. Schumpeter, Capitalism, Socialism and Democracy (New York: Harper & Row,
1942), p. 84.
12
Charles J. Whalen, “Today’s Hottest Economist Died 50 Years Ago,” Business Week,
December 11, 2000, at
http://bear.cba.ufl.edu/christensen/ecp4330/Handouts/TodaysHottestEconomistdied50yearsago
.pdf#search='Age%20of%20Schumpeter'. See also Frank Rose, “The Father of Creative
Destruction,” Wired, March 2002 at
http://www.wired.com/wired/archive/10.03/schumpeter.html.
13
For an extensive summary of dynamic competition theories and references, see Jerry Ellig and
Daniel Lin, “A Taxonomy of Dynamic Competition Theories,” in Jerry Ellig (Ed.), Dynamic
Competition and Public Policy (New York: Cambridge University Press, 2001), pp. 16-44.
14
Richard R. Nelson, “The Tension Between Process Stories and Equilibrium Models: Analyzing
the Productivity-Growth Slowdown of the 1970s,” in Richard N. Langlois, Economics as a
Process: Essays in the New Institutional Economics (Cambridge: Cambridge University Press,
1986).
15
Friedrich Hayek, “Competition as a Discovery Procedure,” in Hayek, New Studies in
Philosophy, Politics, and Economics (Chicago: University of Chicago Press, 1978) pp. 179-90;
Israel Kirzner, “The Perils of Regulation: A Market Process Approach,” in Discovery and the
Capitalist Process (University of Chicago Press, 1985) pp. 119-49; Kirzner, Competition and
Entrepreneurship (Chicago: University of Chicago Press, 1973).
16
W. Brian Arthur, “Competing Technologies, Increasing Returns, and Lock-In by Historical
Events,” Economic Journal 97, 1989, pp. 642-665.
17
Jay Barney, “Competence Explanations of Economic Profits in Strategic Management: Some
Policy Implications,” in Ellig, (Ed.), Dynamic Competition and Public Policy, 2001, pp. 45-64.
18
Clifford Winston, “Economic Deregulation: Day of Reckoning for Microeconomists,” Journal
of Economic Literature 31, September 1993, pp. 1263-89; Clifford Winston, “U.S. Industry
Adjustment to Economic Deregulation,” Journal of Economic Perspectives, 12:3 (Summer
1998) pp. 89-110; Jerry Ellig, “Railroad Deregulation and Consumer Welfare,” Journal of
Regulatory Economics, 21:2 (2002) pp. 143-67.
19
Steven Morrison and Clifford Winston, The Evolution of the Airline Industry (Washington:
Brookings, 1995), p. 21.
20
Steven Morrison and Clifford Winston, The Economic Effects of Airline Deregulation
(Washington: Brookings, 1985); George W. Douglas and James C. Miller III, “Quality
Competition, Industry Equilibrium, and Efficiency in the Price-Constrained Airline Market,”
American Economic Review, 64:4, September 1974, pp. 657-669.
21
Robert W. Crandall and Jerry Ellig, Economic Deregulation and Customer Choice (Mercatus
Center, George Mason University, 1997), p. 45. Available at
http://www.mercatus.org/regulatorystudies/article.php/839.html.
22
Ibid., p. 29.
23
See, e.g., http://www.wges.com/gas/res/new.php?ldc=WGV.
24
Clifford Winston, Thomas M. Corsi, Curtis M. Grimm, and Carol A. Evans, The Economic
Effects of Surface Freight Deregulation, (Washington, DC: Brookings, 1990), Table 3-4 and p.
28; Interstate Commerce Commission, The US Motor Carrier Industry Long After
Deregulation, 1992.
25
C.C. Barnekov and Andrew Kleit, “The Efficiency Effects of Railroad Deregulation in the
United States,” International Journal of Transport Economics, 17, 1990, p. 34.
26
Winston, Corsi, Grimm, and Evans, The Economic Effects of Surface Freight Deregulation,
1990.
27
Balhoff and Rowe, Municipal Broadband, 2005, pp. 20-21.
28
Ibid., p. 43.
29
Ibid., p. 46.
30
Ibid., p. 51.
31
Ibid., p. 22. Since the first quarter of 2003, the percentage of households using DSL more than
doubled, from 6 percent to 13.8 percent. The percentage using cable modem increased from
10.5 percent to 18.3 percent.
32
In the tables and figures, the “A” in “ADSL” stands for “asynchronous, which means that
upstream and downstream data speeds are not necessarily the same. Most service marketed as
“DSL” is actually ADSL. A relatively small number of DSL lines offer the same speed in both
directions. These are included in the “other wireline” category due to the way the FCC
classified the data prior to 2005. These synchronous DSL lines account for fewer than 1
percent of all high-speed lines.
33
Ibid., p. 22.
34
Federal Communications Commission, Wireline Competition Bureau, Industry Analysis and
Technology Division, High-Speed Services for Internet Access: Status as of December 31,
2005 (July 2006), p., 2, available at http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-
266596A1.pdf
35
Ibid., Table 1.
36
Compare data for Verizon Fios and the cable modem services in the “Private Systems” chart in
the Appendix A.
37
Calculated from data in FCC, High-Speed Services for Internet Access (July 2006), Table 15.
38
U.S. Government Accountability Office, Telecommunications: Direct Broadcast Satellite
Subscribership Has Grown Rapidly, but Varies Across Different Types of Markets, Report #
GAO-05-257, April 2005, p. 31. For references to numerous other studies, see Jerry Brito and
Jerry Ellig, “Video Killed the Franchise Star: The Consumer Cost of Cable Franchising and
Proposed Policy Alternatives,” Social Science Research Network Working Paper, March 2006,
available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=893606.
39
U.S. Government Accountability Office, Telecommunications: Wire-Based Competition
Benefited Consumers in Selected Markets Report # GAO-04-241, February 2004, pp. 15-16.
40
FCC, High-Speed Services for Internet Access, July 2005, Table 12.
41
Ronald Rizutto, A Financial Assessment of Municipal Telecommunications Investments in the
United States, Reason Foundation working paper, 2006, pp. 5-7.
42
Balhoff and Rowe, Municipal Broadband, 2005, p. 36.
43
Ibid., pp. 38-39.
44
Calculated from figures in Balhof and Rowe, Municipal Broadband, 2005, p. 39.
45
Ibid., p. 91.
46
Ibid., pp. 90-91.
47
Ibid., p. 93.
48
Bureau of Labor Statistics CPI series for landline long-distance charges, wireless telephone
services, Internet services and electronic information providers, and telephone hardware,
calculators and other consumer information items converted to real CPI using CPI-U for all
items.
49
50
http://www.verizonwireless.com/b2c/mobileoptions/broadband/index.jsp?action=
broadbandAccess, accessed 10/28/05.
51
52
See Appendices for statistics and data sources.
53
See Stephen Margolis, “Path Dependence and Public Policy: Lessons from Economics,”
manuscript, North Carolina State University, January 2005, and references cited therein.
54
S.J. Liebowitz and Stephan E. Margolis, “Path Dependence, Lock-In, and History,” Journal of
Law, Economics, & Organization, 11:1, 1995, pp. 205-26.
55
Margolis, “Path Dependence and Public Policy,” pp. 29-35.
56
Scott Wallstein, “Broadband Penetration: An Empirical Analysis of State and Federal
Policies,” AEI-Brookings Joint Center for Regulatory Studies Working Paper No. 05-12, June
2005, p. 3.
57
See Table 1.
58
“Wireless Battle Ahead?” Dayton Business Journal, November 15, 2005.
59
Most recently, this point has been emphasized in analyses of the FCC’s Total Element Long-Run
Incremental Cost standard for regulating prices that the incumbent telephone companies charge
competitors who lease elements of their networks. See David N. Mandy, “Pricing Network
Elements When Costs Are Changing,” Telecommunications Policy, 26, 2002, pp. 53-67; David M.
Mandy, “TELRIC Pricing With Vintage Capital,” Journal of Regulatory Economics, 22:3,
November 2002, pp. 215-249; David M. Mandy and William W. Sharkey, “Dynamic Pricing and
Investment from Static Proxy Models,” OSP Working Paper Series, Paper #40, September 2003,
pp. 8-9. Available at http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-238934A2.pdf.
60
Balhoff and Rowe, Municipal Broadband, pp. 86-88.
61
Calculated by Aswath Damodaran, available at
http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/Betas.html
62
Ronald Rizutto, A Financial Assessment of Municipal Telecommunications, p. 6.
63
Ibid., pp. 17-18.
64
Ibid., p. 8.
65
Balhoff and Rowe, Municipal Broadband, p. 34.
66
Ibid., p. 35.
67
Ibid., p. 39.
68
Ibid., p. 43.
69
Ibid., p. 48.
70
http://www.econlib.org/library/Enc/RisklessSociety.html.
71
Frank Knight, Risk, Uncertainty and Profit (Chicago: University of Chicago Press, 1971), Ch.
7; Ludwig von Mises, Human Action, 3d Revd. Ed. (Chicago: Contemporary Books, 1966), pp.
105-114.
72
Michael Sciannamea, “Intel Jumping the Gun on WiMAX?,” thewimaxweblog, available at
http://wimax.weblogsinc.com/
73
“Wireless Battle Ahead?,” Dayton Business Journal, November 15, 2005.
74
Knight, Risk, Uncertainty and Profit, pp. 241-43; 291-98.
75
“Wi-Fi Plan May Give City Digital Edge,” Milwaukee Sentinel, October 12, 2005.
76
“This Week in Wireless Cities,” CNET News, October 7, 2005.
77
“Wireless Philadelphia Found Deal Hard to Refuse, but Risks Abound,” Associated Press,
October 6, 2005.
78
Michael Liedke, “Google Still Faces Hurdles Before Connecting San Francisco to Web,”
Associated Press, October 4, 2005; Verne Kopytoff and Ryan Kim, “Google Offers S.F. Wi-Fi
– For Free,” San Francisco Chronicle, October 1, 2005.
79
“City Seeking Network Partner,” Houston Chronicle November 15, 2005.
80
“Google Details Mountain View WiFi Plans,” WebProNews, November 11, 2005.
81
Wallstein, “Broadband Penetration,” p. 5.
82
Ibid., pp. 11-12.
83
See Brito and Ellig, “Video Killed the Franchise Star,” pp. 7-13.
84
47 U.S.C. § 541(a)(1) (2000).
85
FCC, High-Speed Services for Internet Access.
86
Federal Trade Commission, Municipal Provision of Wireless Internet, Staff Report (Sept.
2006), p. 42, available at
http://www.ftc.gov/os/2006/10/V060021municipalprovwirelessinternet.pdf.
Reason Foundation
3415 S. Sepulveda Blvd., Suite 400
Los Angeles, CA 90034
310/391-2245
310/391- 4395 (fax)
www.reason.org

be0b90eed394fb4cbad3233ab06d3c66

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

be0b90eed394fb4cbad3233ab06d3c66

Transféré par

Droits d'auteur :

Formats disponibles

November 2006

A Dynamic Perspective on Government

Reason Foundation’s nonpartisan public policy research promotes

Reason Foundation is a tax-exempt research and education organiza-

Copyright © 2006 Reason Foundation. Photos used in this publication

A Dynamic Perspective on Government

 Competition: Unlike a monopolist, an enterprise that faces competition cannot count on a

 Performance Competition: Competitive businesses seek to continually improve performance—

 Obsolescence: Wireless technology improves rapidly, and as a result capital investment

Several types of government-sponsored broadband initiatives have evoked discussion in recent

What Is Dynamic Competition and

Dynamic competition is a powerful force in the marketplace that makes a substantial

New Issues for Decisionmakers

D ynamic competition complicates the assessment of government-sponsored broadband

financial assumptions underlying Philadelphia’s proposed wireless network imply a residential

Figure 1: Residential and Small Business High-Speed Lines

Source: FCC High-Speed Services Report (July 2006), Table 1.

Table 1: Subscriptions to High-Speed Services (Year-End)

Table 2: Market Share for High-Speed Services (Year-End)

Figure 2: Changing Market Shares

Source: FCC High-Speed Services Report (July 2006), Table 1.

more than 35 percent of residential subscribers, or about 10 percent of households. A fiber

 How fast is it?

An effective government-owned competitor must be prepared to offer a

An effective government-owned competitor must be prepared to offer a price/performance

Figure 3: Real Consumer Price Index, Internet Services

Source: Bureau of Labor Statistics

Figure 4: Monthly Consumer Cost of Internet Access, 1998

Figure 5: Residential Internet Speeds Available in 1998 (kilobytes per second)

If recent experience is a guide, government broadband operations will need to be prepared to

Government broadband operations will need to be prepared to continually improve in the

D. Technological Change and Lock-In

If policymakers choose the “wrong” technology, however, increasing returns or network

Monopolized government enterprises (and traditional regulated, investor-owned utilities as well)

In a dynamically competitive market, networks become obsolete faster. Technology improves

Table 3: Utility Risk Levels

Figure 6: Utility Risk Levels

Data Source: Aswath Damodaran, http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/Betas.html

Development of WiMAX technology may be one example of uncertainty faced by government

 Publicly available, audited financial statements prepared according to Generally Accepted

“Free” Networks: Beware of Geeks

 Who are the competitors?

Private and Public Broadband Providers

Government Systems Name Monthly Fee Download Speed Price/

Government Systems Name Monthly Fee Download Speed Price/

Government Systems Name Monthly Fee Download Speed Price/

Government Systems Name Monthly Fee Download Speed Price/

Notes on Sources and Calculations

About the Author

Related Reason Studies

The Municipal Broadband Compact, September 2005,

Questions Public Officials Should Ask About Government Broadband Services

Spectrum Privatization: Removing the Barriers to Telecommunications Competition, by David

On the Frontier of Deregulation: New Zealand Telecommunications & the Problem of

Vous aimerez peut-être aussi

Competition: Unlike a monopolist, an enterprise that faces competition cannot count on a

Performance Competition: Competitive businesses seek to continually improve performance—

Obsolescence: Wireless technology improves rapidly, and as a result capital investment

How fast is it?

Publicly available, audited financial statements prepared according to Generally Accepted

Who are the competitors?